An Update on the Immunotherapy for Oropharyngeal Squamous Cell Carcinoma

Oropharyngeal squamous cell carcinoma (OPSCC) is an uncommon malignancy worldwide. Remarkably, the rising incidence of OPSCC has been observed in many developed countries over the past few decades. On top of tobacco smoking and alcohol consumption, human papillomavirus (HPV) infection has become a major etiologic factor for OPSCC. The radiotherapy-based or surgery-based systemic therapies are recommended equally as first-line treatment, while chemotherapy-based strategy is applied to advanced diseases. Immunotherapy in head and neck squamous cell carcinoma (HNSCC) is currently under the spotlight, especially for patients with advanced diseases. Numerous researches on programmed death-1/programmed death-ligand 1 checkpoint inhibitors have proven beneficial to patients with metastatic HNSCC. In 2016, nivolumab and pembrolizumab were approved as the second-line treatment for advanced metastatic HNSCC by the USA Food and Drug Administration. Soon after, in 2019, the USA Food and Drug Administration approved pembrolizumab as the first-line treatment for patients with unresectable, recurrent, and metastatic HNSCC. It has been reported that HPV-positive HNSCC patients were associated with increased programmed death-ligand 1 expression; however, whether HPV status indicates different treatment outcomes among HNSCC patients treated with immunotherapy has contradicted. Notably, HPV-positive OPSCC exhibits a significantly better clinical response to primary treatment (i.e., radiotherapy, surgery, and chemotherapy) and a more desirable prognosis compared to the HPV-negative OPSCC. This review summarizes the current publications on immunotherapy in HNSCC/OPSCC patients and discusses the impact of HPV infection in immunotherapeutic efficacy, providing an update on the immune landscape and future perspectives in OPSCC.


INTRODUCTION 2 HPV AND OPSCC
HPV is a small non-enveloped, circular, double-stranded DNA virus with epithelial tropism and commonly transmits by sexual contact. It has been reported that HPV infection was attributed to around 20%~60% of OPSCC worldwide (9,21). Among 200 identified genotypes of HPVs, genotype HPV-16 accounts for over 80% of HPV-positive OPSCC, followed by HPV-18, 31, 33, 35, which are well-known high-risk HPVs (9,22). E6 and E7, two early viral proteins expressed by high-risk HPV, are mainly involved in developing and maintaining the transformed phenotype of HPV-induced cancers (23). Specifically, the oncoprotein E6 degrades the tumor suppressor p53 and helps escape cell death (21,24). E7 binds to the retinoblastoma proteins (pRb), promoting the E2F/pRb complex dissociation and releasing E2F, which stimulates the cell re-entering S-phase, leading to escape from oncogene-induce senescence (21,25). Besides, E6, E7, and E6/E7 contribute to the maintenance of cancer phenotype, epigenetic regulation, microRNAs, DNA damage response, genetic instability, angiogenesis, immune system modulation, telomerase activity via a variety of molecules/pathways (25). Although playing a less crucial role, other early proteins, including E1, E2, E4, and E5, participate in completing the viral cell cycle (9,21).
The 7 th edition of the Union for International Cancer Control and American Joint Committee on Cancer (UICC/AJCC) staging system failed to differentiate the impact of HPV infection on survival (hazard consistency) between stages and lost the capability in predicting features of any stage (37). Hence, the latest released 8 th edition has introduced apparent modifications in a new staging algorithm to categorize OPSCC patients into two different systems regarding HPV status. We summarize the difference between the 7 th and 8 th edition of UICC/AJCC staging systems of OPSCC in the Supplementary Table. The College of American Pathologist Guidelines recommended that p16 expression tested by immunohistochemistry is a feasible and reliable surrogate marker for diagnosing HPV-positive OPSCC (38,39). It is worth noting that discordance between p16 staining and HPV status (e.g., patients with p16 INK4a -positive/HPV DNA-negative) has been observed and related to differentiated survival (40,41). Therefore, we shall be aware that p16 INK4a alone may not be the best biomarker for prognosis prediction. Alternative and/or complementary biomarker, such as DNA and RNA in situ hybridization and other molecular HPV tests, is urgently needed (37,42).

IMMUNOTHERAPY
A significant breakthrough has been achieved in cancer immunotherapy, making it an important weapon in fighting cancer (43)(44)(45)(46). According to the Cancer-Immunity Cycle proposed by Chen et al., cancer cells can be effectively eradicated by the immune cells via a stepwise process which starts with a successful initiation of cancer immune recognition and accumulation of adaptive immune responses, to cancer cell elimination eventually (47). Nevertheless, the Cancer-Immunity Cycle does not always work desirably in cancer patients; for instance, T-cell-mediated attack might fail to activate due to the suppression by some factors in the tumor microenvironment (48). Each step in the Cycle acts as a potential strategy for cancer immune escape and an eligible target for treatment (49). Among those steps, PD-1/PD-L1 checkpoint axis is most widely studied, which prevents the over-activation of T cells from damaging normal tissues and leads to the potential of tumor immune escape (12). In the past decade, blocking the PD-1/PD-L1 axis by monoclonal antibodies to overcome the immune suppressive signals in cancer patients and promote the reactivation of antitumor response has been well-established as an efficient treatment in a broad range of cancer types (including but not limited to lung cancer, breast cancer, head and neck cancer, pancreatic cancer, and prostate cancer) (50-54).  Table 1.

Immunotherapy-Combined Treatments in HNSCC/OPSCC
SoC of HNSCC, including surgery, radiotherapy, chemotherapy, and targeted therapy, have been well-studied and widely applied in clinics with a proven impact (72). Nevertheless, the efficacy of SoC has reached a plateau, and a novel therapeutic modality is urgently needed. After the notable success of PD-1/PD-L1 achieved in treating advanced HNSCC, researchers have devoted their passion to explore the potential of immunotherapy-combined treatments.
Lenvatinib is a tyrosine kinase inhibitor of several VEGF receptors and could modulate immune suppression in the tumor micro-environment by angiogenesis inhibition. The effectiveness of pembrolizumab in combination with lenvatinib in patients with HNSCC has been supported by a phase II trial (NCT02501096) (72)  Besides, the impact of ICI therapy combined with chemotherapy was investigated in a wide range of solid tumors. Some of these trials (KEYNOTE-189, 355, 361, 407, 590, 826) observe statistically significant survival benefits (longer PFS, higher estimated rate of overall survival at 12 months) in patients with corresponding cancers (73)(74)(75)(76)(77)(78), suggesting adding pembrolizumab to standard chemotherapy for cancer treatment. As described above, the KEYNOTE-048 trial reported that pembrolizumab+chemotherapy improved HNSCC patients' OS versus cetuximab with chemotherapy (13.0 vs. 10.7 months, P=0.0034), while pembrolizumab monotherapy was noninferior to cetuximab with chemotherapy (11.6 vs. 10.7 months). When stratifying the patient population with CPS score, both arms show survival benefits compared to cetuximab with chemotherapy. However, 85% of the R/M HNSCC patients in the pembrolizumab with chemotherapy group suffered grade 3 or worse all-cause adverse events (AEs), while 55% of the patients in the pembrolizumab monotherapy group endured AEs. It indicates that when clinicians decide on monotherapy or combined therapy, the toxicity should be considered in the clinical settings.

Dual Immune Checkpoint Blockade Therapy in HNSCC/OPSCC
The desirable effect of single-agent immunotherapy sparks research into the combination of anti-CTLA-4 and anti-PD-1 therapies (79). Ipilimumab, a monoclonal antibody medication targeting CTLA-4, has been approved for application in melanoma, advanced high-risk renal cell carcinoma, colorectal cancer (80)(81)(82). A phase II, open-label, randomized clinical trial (NCT02919683) evaluating the effect of nivolumab (N arm) or nivolumab+ipilimumab (N+I arm) in HNSCC patients before surgical resection showed that both arms had a favorable response, whereas RR in the N+I arm was better in the RECIST manner (N+I arm: 38%, N arm: 13%) (83).
Among these novel therapies, researchers in the field of HNSCC, including OPSCC, start to explore the effect of IDO1 inhibition therapy and toll-like receptor 8 (TLR8) agonists therapy through clinical trials. The results of the phase Ib study (NCT02471846) were disappointing that the combination of navoximod (IDO1 inhibitors) and atezolizumab (anti-PD-L1 agent) failed to improve clinical benefit among patients with various solid tumors, including HNSCC (89). However, the results of the Active8 study were encouraging that TLR8 agonists might prolong survival among HPV-positive HNSCC patients compared to HPV-negative patients (PFS: 7.8 versus 5.9 months; HR, 0.58; P = 0.046; OS: 15.2 versus 12.6 months; HR:0.41; P = 0.03) (59).
In all, immunotherapy provides a promising future, but the application in the treatment of OPSCC is still lacking. Researchers should take steps to discover more information in this field.

Immunotherapy in HPV-Positive OPSCC
HPV is a solid causative agent in the formation and progression of OPSCCs, making viral neoantigens an attractive target for therapeutic immunization. Tumor vaccine aims to reduce tumor burden and control tumor recurrence by stimulating both humoral and cellular immune response, offering an immune activation strategy (90). Current HPV-positive therapeutic vaccines are depicted in Figure 1.
Given the essential roles E6 and E7 play in HPV-positive cancers, they are usually selected as targets for a therapeutic vaccine. A phase I study (ACTRN12618000140257) assessed the safety, tolerability, and immunogenicity of an HPV E6/E7 vaccine (AMV002) in patients with HPV-positive OPSCC (91). The vaccine-induced RR was 83.3% (10 of 12). In addition, a phase Ib/II trial (NCT03162224) evaluating the safety and efficacy of MEDI0457 (DNA vaccine targeting HPV-16/18 E6/ E7 antigens accompanied with an IL-12 adjuvant) plus durvalumab in HPV-positive R/M HNSCC patients is underway (92).
On top of various vaccines, Ramireddy et al. showed that tumor membrane vesicle (TMV) vaccine comprises glycolipid-anchored immuno-stimulatory molecules GPI-B7-1 and GPI-IL-12, magnified the efficacy of anti-PD1 antibodies and inhibited tumor growth, and thus improved the survival of mice with stage VII SCC (93). In addition, the early expressed HPV E5 protein has gained attention recently (94). By amplifying EGFR, HPV E5 protein promoted cell proliferation and invasion through Ras-ERK1/2, PI3K-AKT, and COX-2 pathways (95). As more signaling pathways are involved in understanding tumor-related events, new target sites are under active investigation.
However, those HPV therapeutic vaccines were neither successful in achieving desired clinical outcomes nor applied to human study. Researchers put forward that one reason might be that patients obtained immune tolerance to antigens due to chronic exposure to the virus for years (96). Therefore, we suggest that improving the immunogenicity of several types of vaccines and ensuring safety and tolerability warrant more future efforts.

Immunotherapy in HPV-Negative OPSCC
Currently, there are few studies on immunotherapy towards HPV-negative OPSCC. DURTRERAD is a randomized phase II trial evaluating feasibility and efficacy of durvalumab (D arm) versus durvalumab and tremelimumab (DT arm) in combination with radiotherapy as primary treatment for locally advanced HPV-negative HNSCC, more than a half being OPSCC (97). In this trial, the DT arm was halted due to increased toxicity. Five among six patients in the DT arm suffered varying degrees of trAEs, with one quitting the cohort because of grade 5 trAE. However, detailed analysis has not been fully reported (97).

Survival Benefits of Immunotherapy in Relation to HPV Status
As mentioned above, the prognosis of OPSCC patients is closely associated with HPV status in the setting of SoC. Nevertheless, HPV status seems to be limited in predicting the prognosis of immunotherapies towards OPSCC patients. A systematic review of clinical trials using immunotherapy, mainly ICIs in HNSCC, did not report any statistically significant difference concerning ORR, stable disease (SD), progressive disease (PD), or OS when patients were classified by HPV status. However, ORR was approaching significance in HPV-positive patients compared to HPV-negative ones (21.2% vs. 15.0%, P = 0.06) (7).
It should be pointed out that the accuracy of existing detection methods would influence the result of these trials. A study held by Miren Taberna et al. Proves that the ICON-S model in 8th AJCC predicts overall survival assessment better in HPV-related OPC patients when using at least two biomarkers to define HPV-causality (HPV-DNA and (p16INK4a or HPV-mRNA) or double positivity for HPV-DNA/p16INK4a) (98). These results call for a clear definition of HPV status in future studies, and competent biomarkers to identify the beneficiaries from immunotherapies.

Biomarkers for Immunotherapy in OPSCC
Cancer biomarkers help evaluate treatment effects, monitor tumor recurrence and predict survival. The application of biomarkers will influence patient outcomes; hence, exploring and discovering novel biomarkers have significant research value.
The prediction value of PD-(L)1 expression measured by immunohistochemistry for prognosis in patients with HNSCC was analytically validated (99). Previous studies evaluating durvalumab treatment effect demonstrated a better anti-tumor response in R/M HNSCC patients with high PD-L1 expression versus those with low or no PD-L1 expression (median OS: 7.1 vs 6.0 months, ORR: 16.2% vs 9.2%) (100). Nevertheless, the cut-off value varies from trials to trials, and researchers need more trials to define a uniform standard when using PD-(L)1 expression as a prognostic marker (100).
Recent efforts to excavate molecular biomarkers through next-generation sequencing provided deeper and broader insights. Genetic and epidemic alterations are involved in the initiation of HNSCC formation. Genetic alterations include the classic mutations in TP53 and CDK2NA and the newly discovered mutations in FBXW7, TP63, IRF6, and NOTCH1 (32). NOTCH signaling pathways were associated with the development of multiple types of tumors, such as hepatocellular carcinoma, T cell leukemias, myeloid leukemia (101)(102)(103). Gianluigi Grilli et al. reported that the activation of the NOTCH pathway improved prognosis in HPV-negative HNSCC patients and suggested that NOTCH1 expression might be a predictive marker for survival in HPV-negative HNSCC (104). Besides, Esposti et al. identified a unique epigenetic feature: hypomethylation in NCAN, NRXN1, COL19A1, SYCP2, RPA2, and SMC1B, related to HPV infection among HNSCCs regardless of the anatomic site (105). Moreover, differentially expressed small non-coding RNA molecules (miRNAs) may also predict survival. Expression of miR-21 was associated with poor cancer-specific survival in HPV-negative tumors (106).

Resistance to Immune Checkpoint Blockade
Most clinical trials evaluating the RR to ICIs in OPSCC patients indicate that less than 15% of patients receiving ICI therapy could achieve durable responses (14,15,107). Evidence supports that failure of immune sensing might contribute to compromised immune function. Two dysfunctional oncogenic pathways, the SOX2-mediated suppression of the IFN-I signaling pathway and the PI3K-mTOR pathway, deprive extracellular glucose and thus exhaust cytotoxic T lymphocytes (CTLs) (108). Pervasive immune suppression is the primary barrier, contributing to the limited beneficial effects of ICIs in OPSCC. Researchers proposed that targeting the IFN-I signaling pathway through IFN-I agonists (e.g., cGAMP), inducing DNA damage by RT or DNA-damage inducing agents (e.g., cisplatin and 5-fluorouracil), and is revitalizing CTLs (e.g., rebuilding a pH-neutralized environment to provide nutrition) might be effective (109)(110)(111)(112). In addition, Zhou L. et al. suggest that epigenetic targeting drugs such as DNA methyltransferase inhibitors, histone deacetylase, and methyltransferase inhibitors may potentially reverse immune suppression in various cancer models (113).

Difficulty in Personalized Immunotherapy Strategy
In the clinical practice of immunotherapy, evaluating patients' immune status could be challenging. As discussed above, feasible biomarkers could guide researchers to identify potential beneficial patients and monitor adverse events. However, it is insufficient to select patients for ICI immunotherapy based on a single parameter without considering other factors given the highly heterogeneous microenvironment in OPSCC. An ex vivo platform, CANscript system, has been proven helpful to for profiling the response of immunotherapy combining chemotherapy (114). Besides, researchers have established several prognosis risk models to indicate immunosuppression state and predict survival in patients based on a set of immune checkpoint-related genes (115,116). It is a pity that, however, existing predictive models lack validation upon large sample size.

CONCLUSION
In summary, classified by HPV status, OPSCC is a heterogeneous disease. The unique TME shaping by HPV status calls for distinct therapeutic approaches. Immunotherapy offers a wide range of therapeutic strategies which will be especially useful in meeting this need. Monotherapy of novel agents has proved effective, while combinations of immunotherapy with conventional therapies and dual immunotherapy are undergoing clinical investigation. Notably, there is an urgent need for a feasible treatment stratification by HPV status. Immunosuppression and lack of desirable biomarkers for personalized therapy are the two significant issues in immunotherapy. More clinical trials are warranted to assess the efficacy of novel immunotherapies based on HPV status.

AUTHOR CONTRIBUTIONS
All authors listed have made a substantial, direct, and intellectual contribution to the work, and approved it for publication.